Wikimpacts — A global climate impact database based on automated information extraction from Wikipedia

Ni Li, Wim Thiery, Shorouq Zahra, Mariana Madruga de Brito, Koffi Worou, Murathan Kurfali, Seppe Lampe, Paul Munoz, Clare Flynn, Camila Trigoso, Joakim Nivre, Jakob Zscheischler, Gabriele Messori

We introduce Wikimpacts 1.0, a new global database on climate impacts developed using natural language processing techniques. Impact data for each event is recorded at three levels: event, national, and sub-national. Categories include the number of deaths, injuries, homelessness, displacements, affected individuals, damaged buildings, and insured or total economic damages.

Extreme climate events like storms, heatwaves, wildfires, floods, and droughts pose serious threats to human society and ecosystems. Impact database can be used in impact forecasting, early warning, and disaster risk management.

This database utilizes the GPT4o large language model for extracting information, following document selection, post-processing, and data consolidation. In this release, we have processed 3,368 Wikipedia articles. This dataset encompasses 2,928 events from 1034 to 2024, featuring 20,186 national and 36,394 sub-national data entries.

TerrestrialClimate ImpactsImpact databaseClimate extremesGPT4oLLM

Name

Version

Citation

Data description

The database is currently available in SQL format and provided as a .db file. The SQL format comprises four components:

Total_Summary: This table provides impact information at the event level and includes 2,928 events. Impact information includes:

• Event_ID which is internal to the database
• Event_Names
• Sources which provides a URL to the relevant Wikipedia article
• Main_Event, e.g. wildfire, tornado, flood and the like
• Hazards resulting from the event, e.g. wind or lightning
• Administrative_Areas_Norm
• Administrative_Areas_GID: the administrative area in Global Impact Database format
• Administrative_Areas_Type: e.g. country, state, town, etc.
• Administrative_Areas_GeoJson: the administrative area in GeoJSON format
• Start_Date_Day as the date of a month
• Start_Date_Month as the number of the month
• Start_Date_Year
• End_Date_Day as above
• End_Date_Month as above
• End_Date_Year
• Total_Deaths_Min: the minimum estimated number of fatalities
• Total_Deaths_Max: the maximum estimated number of fatalities
• Total_Deaths_Approx: 1 if the number of fatalities is approximate, 0 if not
• Minimum, maximum and approximate values for Total_Injuries_*, Total_Affected_*, Total_Displaced_*, Total_Homeless_*, Total_Buildings_Damaged_*, Total_Insured_Damage_*, Total_Damage_*
• Values for Total_Insured_Damage_* and Total_Damage_* have information on unit (e.g. USD), adjustment for inflation, and the year of inflation adjustment

Instance_*: There are 8 tables that provide impact information at the national level, with a total of 20,186 entries. Each of these tables provide data for Event_ID, Administrative_Areas_Norm, Administrative_Areas_GID, Administrative_Areas_Type, Start_Date_Day, Start_Date_Month, Start_Date_Year, End_Date_Day, End_Date_Month, End_Date_Year, Num_Min, Num_Max and Num_Approx. The tables are:

• Instance_Per_Administrative_Areas_Affected
• Instance_Per_Administrative_Areas_Buildings_Damaged
• Instance_Per_Administrative_Areas_Damage
• Instance_Per_Administrative_Areas_Deaths
• Instance_Per_Administrative_Areas_Displaced
• Instance_Per_Administrative_Areas_Homeless
• Instance_Per_Administrative_Areas_Injuries
• Instance_Per_Administrative_Areas_Insured_Damage

Specific_Instance_Per_Administrative_Area_*: These 8 tables detail impact information at the sub-national level, with a total of 36,394 entries. The tables are structured as the previous ones.

Geo_Json_Obj: This table contains all GeoJSON objects for geolocation information within the database.

Comments

Gabriele Messori is also affiliated with

• Swedish centre for impacts of climate extremes (climes), Uppsala University, Uppsala, Sweden
• Department of Meteorology and Bolin Centre for Climate Research, Stockholm University, Stockholm, Sweden

This dataset is published under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).

Publisher

Bolin Centre Database

License

First name

Gabriele

Last name or organisation

Messori

Email address

gabriele.messori@geo.uu.se

Address

Department of Earth Sciences, Uppsala University

Postal code

SE-75236

City

Uppsala

Country

Sweden

GCMD science keyword

Earth science > Human dimensions > Natural hazards

GCMD location

Continent

Dataset language

English

DOI

Time

Versions

• 1.0.final 2026-04-05T12:14:05.048+00:00
• 1.0 2025-06-11T09:37:23.288+00:00